Active Embedding Search via Noisy Paired Comparisons

Authors: Gregory Canal, Andy Massimino, Mark Davenport, Christopher Rozell

ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We use simulated responses from a real-world dataset to validate our strategies through their similar performance to greedy information maximization, and their superior preference estimation over state-of-the-art selection methods as well as random queries.
Researcher Affiliation Academia 1School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia, United States.
Pseudocode Yes Algorithm 1: Pairwise search with noisy comparisons
Open Source Code Yes Code available at https://github.com/siplab-gt/pairsearch
Open Datasets Yes We constructed an item embedding of the Yummly Food-10k dataset of (Wilber et al., 2015; 2014), consisting of 958,479 publicly available triplet comparisons assessing relative similarity among 10,000 food items.
Dataset Splits No The paper describes the simulation setup and generation of random points for trials but does not specify training, validation, or test dataset splits with percentages or counts, nor does it reference predefined splits.
Hardware Specification No The paper does not provide specific details about the hardware used for running the experiments, such as GPU/CPU models or memory specifications.
Software Dependencies No The paper mentions 'Stan Modeling Language' (Carpenter et al., 2017) but does not provide a specific version number for it or any other software dependencies.
Experiment Setup Yes In each simulation trial, we generate a point W uniformly at random from the hypercube [ -1, 1]d and collect paired comparisons... We use the Stan Modeling Language... For each scheme we optimized the value of k0 over the set of training triplets via maximum-likelihood estimation... d = 4.